The Countervailing Effects of Community Supervision: Evidence from a Natural Experiment

• In Event: 369 - Regular Session. Criminology

Abstract

While there has been dramatic growth in the number of individuals under community supervision in the United States, its effectiveness in reducing recidivism and incarceration rates remains unclear. On the one hand, community supervision promises to rehabilitate offenders by providing structure and monitoring to deter criminal behavior. On the other, it may widen the net of social control because of high rates of violations, and as a consequence, high rates of incarceration. This study seeks assess these competing views by investigating the causal effects of community supervision on recidivism. Specifically, it assesses whether the addition of probation to a jail sentence reduces reoffending or deepens involvement with the criminal justice system. It draws on original data on a sample of offenders (n=112,660) convicted of a felony in the state of Michigan between 2003 and 2006. It leverages the random assignment of criminal cases to judges and relies on an instrumental variables design to estimate the causal effects of community supervision. The results indicate countervailing effects of community supervision, such that for the marginal offender, being sentenced to jail without community supervision increases the probability of arrest, increases the probability of new felony convictions, but reduces the probability of being incarcerated. Results show that community supervision reduces reoffending for some offenders, but those gains are undercut by its “back-end net-widening” effects of surveillance, revocation due to technical violations, and incarceration.

Authors

• Anh P. Nguyen, University of Michigan-Ann Arbor

• David J. Harding, University of California-Berkeley

• Jeffrey Morenoff, University of Michigan

• Shawn Bushway, University at Albany-SUNY